CVE-2025-46153
September 25, 2025
PyTorch before 3.7.0 has a bernoulli_p decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallback_random=True.
Affected Packages
pytorch (CONDA):
Affected version(s) >=0.1.12 <2.7.0Fix Suggestion:
Update to version 2.7.0https://github.com/pytorch/pytorch.git (GITHUB):
Affected version(s) >=v0.1.1 <v2.7.0Fix Suggestion:
Update to version v2.7.0torch (PYTHON):
Affected version(s) >=0.1 <2.7.0Fix Suggestion:
Update to version 2.7.0Related Resources (5)
Do you need more information?
Contact UsCVSS v4
Base Score:
6.9
Attack Vector
NETWORK
Attack Complexity
LOW
Attack Requirements
NONE
Privileges Required
NONE
User Interaction
NONE
Vulnerable System Confidentiality
LOW
Vulnerable System Integrity
NONE
Vulnerable System Availability
NONE
Subsequent System Confidentiality
NONE
Subsequent System Integrity
NONE
Subsequent System Availability
NONE
CVSS v3
Base Score:
5.3
Attack Vector
NETWORK
Attack Complexity
LOW
Privileges Required
NONE
User Interaction
NONE
Scope
UNCHANGED
Confidentiality
LOW
Integrity
NONE
Availability
NONE
Weakness Type (CWE)
Inefficient CPU Computation
EPSS
Base Score:
0.05
